42 Robotics Roles Posted, Median Salary $230,000 at Atoms

From Stealth to Scale: Atoms' Public Reveal and Robot Deployments

Travis Kalanick broke an eight-year silence on a Friday in March 2026, appearing on the TBPN podcast and publishing a 1,700-word manifesto that renamed his holding company City Storage Systems to Atoms and declared a mandate to deploy autonomous robots across food production, mining, and transportation. The reveal triggered an immediate hiring surge, with 42 open roles listed with a median salary of $230,000 and nine added in the past seven days, and arrived as rivals from Boston Dynamics to Tesla pursue their own physical-AI initiatives in the same three verticals.

The holding company he had run since 2018, parent of the ghost-kitchen operator CloudKitchens, was renamed Atoms. Its mandate: build what Kalanick calls a "wheelbase for robots" in those sectors. Not humanoids. Not general-purpose machines. Specialized, wheeled systems designed for high-cycle industrial environments where the economics of automation are already legible.

The reveal was not a product demo. It was a structural argument. Kalanick's framework treats the physical world through a computer-science lens: if processors manipulate bits, manufacturing manipulates atoms; if data centers store bits, real estate stores atoms; if networks move bits from point A to point B, transport and logistics move atoms. "Treating atoms like bits, digitizing the physical world," he said on TBPN. The company's website now lists three subcategories: Atoms Food ("infrastructure for better food"), Atoms Mining ("more productive mines"), and Atoms Transport ("wheelbase for robots"), each intended to run on a shared autonomy stack rather than bespoke one-offs.

That stack has a proving ground. CloudKitchens operates hundreds of ghost-kitchen locations globally, and its food robotics division, Lab37 in Pittsburgh, is led by Eric Meyhofer, former head of Uber's self-driving unit. Lab37's Bowl Builder system automates up to 40 percent of meal-preparation workflows, SiliconANGLE reported, giving Atoms a controlled environment to refine coordination algorithms before deploying to mines or logistics yards. Kalanick described the food ambition bluntly: do "to the kitchen what Uber did to the car" — make prepared meal delivery cost-efficient enough to approach grocery-shopping prices.

TechCrunch reported that Kalanick disclosed he is already the largest investor in Pronto, an autonomous-vehicle startup focused on industrial and mining sites founded by former Uber colleague Anthony Levandowski. Atoms is now moving to acquire Pronto outright. That deal would hand Atoms an autonomy stack already tested in haul-truck environments where Caterpillar and Komatsu fleets run hundreds of autonomous trucks with documented productivity gains and reduced safety incidents.

Capital is not the constraint. City Storage Systems raised more than $1.25 billion and, as CNBC reported, carried a $15 billion valuation in 2021. Board data shows Atoms already listing 42 open roles with a salary band of $62,000–$360,000 (median $230,000), including a Vice President of Global Controller at $300,000–$360,000 and a Staff Machine Learning Engineer at $273,000–$345,000. Nine roles were added in the past seven days alone. The hiring signal is unambiguous: the stealth period funded the infrastructure; the public phase funds the deployment.

Why now? Kalanick argued on TBPN that AI in the physical world is "way harder" than in the digital world because "there's just way less training data." The eight-year stealth window was spent building CloudKitchens and generating data in ghost kitchens, simulation, and Pronto haul-truck runs while the venture narrative obsessed over humanoids. Figure AI raised $675 million for humanoid manufacturing robots in Q1 2025; Physical Intelligence secured $400 million for cross-platform AI controllers. Atoms is pursuing the inverse strategy: domain-specific hardware optimized for constrained environments with measurable ROI. The bet is that "gainfully employed robots" (wheeled, task-specific, industrially scaled) reach payback periods that general-purpose platforms cannot.

The reveal also reframes Kalanick's own history. He created a self-driving division at Uber in 2015; the subsequent Waymo lawsuit and Levandowski's criminal conviction became a cautionary tale about speed without safety validation. In a rare 2025 interview, Kalanick expressed regret that Uber abandoned its self-driving program. Atoms is the second act — same founder, same obsession with moving atoms, different risk calculus. "I left Uber in 2017 heartbroken," he wrote in the manifesto. "I bled, but I did not perish. I got back up and fought my way back into the arena, back to my calling. Back to building."

The arena is now public. The robots are deploying. The hiring surge has begun. And the competitors, including Boston Dynamics, Tesla Optimus, Foxconn, and a wave of Chinese robotics firms that represented nearly three in five CES 2026 exhibitors, are watching a new entrant with $1.25 billion in dry powder, a live data engine in food, and a mining autonomy stack incoming via Pronto. The physical-AI race just got a heavily capitalized, operationally grounded participant that refuses to build humanoids.

The Workforce Build-Out

Atoms' job board tells the story in numbers the press releases don't. A single LinkedIn listing for a robotics-focused position drew over 200 applicants as of July 2026. Zero G Talent reported that the company posted nine new roles recently, spanning vice-presidential finance leadership at $300,000–$360,000, such a role at $273,000–$345,000, and infrastructure product management alongside high-performance computing network engineering both banded at $224,000–$284,000.

The hiring velocity mirrors a market that has rewritten its own rules. Since late 2022, global AI job advertisements jumped roughly two-thirds. In 2024, postings requiring AI skills surged 61 percent year-over-year while overall job ads grew 1.4 percent. Nearly one in four new tech roles now explicitly seeks AI capabilities. AI positions represent roughly one-fifth of all tech postings — more than double their 2022 share. Machine learning engineer listings grew 40 percent from 2024 to 2025, stacking on a 78 percent increase the year before. Robotics engineering roles rose 11 percent, a direct signal that AI is moving from screens into the physical world.

Atoms sits at that intersection. The company's careers page frames the mission bluntly: "The next great unlock is autonomy in the world of atoms: growing, mining, manufacturing, and moving." Its LinkedIn job description reads: "Atoms builds Physical AI — real-world robots for the industries that move civilization forward, starting with food, mining, and transport." CEO Travis Kalanick puts the constraint in starker terms: "The constraint on human progress is our ability to automate the physical world."

The talent shortage is measurable. Three-quarters of large companies report a severe AI talent gap even as 93 percent call AI crucial to their future. That scarcity commands a premium — AI roles pay 20–30 percent above comparable non-AI positions. Atoms' compensation bands reflect the pressure. Zero G Talent found the staff machine learning engineer role tops out at $345,000. According to Zero G Talent, the VP global controller posts at $360,000 across San Francisco, New York, and Los Angeles simultaneously — a geographic spread that signals the company is building distributed leadership, not a single headquarters team.

Competitors are moving too. Data shows Boston Dynamics added five roles in the same seven-day window, with a median salary of $140,000 and a band from $21,000 to $275,000 across 16 total openings. Their latest postings target reinforcement learning research, human-robot interaction UX, and manufacturing engineering, all disciplines Atoms also needs. The overlap is not coincidental. Both companies are chasing the same scarce pool of engineers who can bridge simulation and hardware, perception and control, fleet orchestration and single-robot reliability.

Atoms' own careers site acknowledges the pipeline problem with an "Emerging talent" track for students and early-career builders, plus internships that "contribute to live products" across robotics, AI, software, manufacturing, operations, and design. A 2025 industry video captured the shifting skill set: "Besides understanding the technicalities of the machines they are operating, robotic engineers are also expected to have programming skills and an understanding of how different technologies can be integrated together." The same source noted: "In this field, technology is evolving every day, so you definitely need to be curious about all the latest technology."

The hiring surge is not a recruiting campaign. It is the visible edge of a capital allocation decision. Atoms is converting the eight-year stealth investment (thousands of employees built under City Storage Systems) into a deployed physical-AI platform across three heavy industries simultaneously. Each role posted is a bet on a specific subsystem: high-performance networking for fleet-scale data ingest, machine learning for perception and planning, product management for the infrastructure layer that makes robots programmable rather than programmable-only. The 200-plus applicants on one listing suggest the market understands the bet. Whether the talent base can scale fast enough to fill 42 roles at median $230,000 while Boston Dynamics, Tesla Optimus, and Foxconn all fish the same waters is the operational question the next quarter will answer.

Where the Money Moves

Persistent labour shortages, escalating cost pressures, and fragile global supply chains are converging to threaten productivity across industrial sectors. The World Economic Forum's 2025 white paper on physical AI frames the problem bluntly: these forces "are converging to threaten productivity, profitability and resilience." Geopolitical instability and market uncertainty have only sharpened the squeeze. At the same time, consumer expectations for speed, customization, and sustainability demand operational flexibility that traditional automation cannot deliver.

The numbers illustrate the scale. By 2023, more than 4 million industrial robots had been installed globally. Amazon alone operates more than 1 million robots across its network, making it the largest user of robotics worldwide. The food and beverage, metal processing, logistics, and discrete manufacturing sectors stand to benefit most — precisely the domains where Atoms has planted its flag. Most manufacturing companies are small- and medium-sized enterprises, and the WEF research emphasizes they stand to gain just as much as large enterprises.

Safety regulations and injury costs add another layer of urgency. Robotics can absorb tasks involving lifting, repetitive motion, or exposure to hazardous environments, common causes of workplace injury. At Amazon sites integrating robotics, incident rates have fallen 15 percent as employees shift into oversight, diagnostics, and continuous-improvement roles. The Shreveport, Louisiana fulfilment centre, an early next-generation deployment, recorded improved workplace safety alongside 25 percent faster delivery and a 25 percent efficiency gain.

Efficiency gains compound when physical AI replaces rigid automation. Foxconn cut deployment time by 40 percent using AI-powered robotics and digital-twin simulation for high-precision tasks such as screw tightening and cable insertion, work previously considered too complex to automate. Real-time adaptive force control and simulation-based deployment reduced operational costs by 15 percent. Cycle times improved 20–30 percent. Force feedback and motion control advances cut error rates by 25 percent. Virtual validation eliminated costly trial-and-error in physical environments. The self-adjusting force and trajectory of AI-driven robotic arms now achieve higher success rates than human workers in complex assembly.

A generative AI foundation model coordinating Amazon's entire mobile robot fleet added another 10 percent fleet travel efficiency. These are not marginal improvements. They rewrite the economics of operations that were previously too variable or cost-prohibitive to automate. As physical AI supports a wider variety of tasks and becomes easier to deploy (requiring fewer specialized skills and less task-specific customization), automation becomes viable across a much broader range of operations.

The WEF's Future of Jobs Report 2025 projects robots and autonomous systems as the leading driver of net job displacement by 2030. But new skilled roles are emerging around supervising intelligent systems, resolving edge cases, and optimizing performance. The shift is already visible: next-generation facilities report 30 percent more skilled roles onsite. Manufacturers who delay risk missing the momentum of rapidly converging technology and economic tipping points. The competitive imperative is no longer theoretical — it is measured in deployment cycles, injury rates, and cost per unit.

Rivals Dig In

Atoms' public reveal and its stated focus on food processing, mining haulage, and transportation logistics landed in a market already mobilizing. Incumbents and well-funded challengers have spent the past 18 months locking in customers, expanding factory capacity, and rewriting product roadmaps. The timing is not coincidental — each move tightens the competitive perimeter around the three verticals Atoms now targets.

Boston Dynamics has made the loudest hardware bet. In June 2026 the company confirmed a $100 million manufacturing facility in Waltham, Massachusetts, a 323,000-square-foot site at 1601 Trapelo Road that would consolidate production of Spot, Atlas, and the warehouse-focused Stretch. Massachusetts committed $25 million in tax credits through its Economic Assistance Coordinating Council, and Boston Dynamics projects up to 1,250 employees at the site by 2033. The investment follows a deployment curve that Hyundai, which acquired Boston Dynamics from SoftBank in 2021, said at CES 2026 has already unloaded more than 20 million boxes worldwide since Stretch launched. DHL Group, an early buyer alongside Gap and H&M when Stretch became commercially available in March 2022, signed a memorandum of understanding in May 2025 to deploy more than 1,000 additional units by 2030. DHL has invested over €1 billion in automation across its contract logistics division in the past three years and now runs more than 7,500 robots, 200,000 smart handheld devices, and nearly 800,000 IoT sensors globally; over 90 percent of its warehouses worldwide have at least one automation or digitalization solution. The logistics giant is also shifting toward co-development with robotics partners rather than relying solely on off-the-shelf systems. UPS, meanwhile, announced in December 2025 a $120 million purchase of roughly 400 unloading systems from Pickle Robot, a direct Stretch competitor. Lidl converted its Stretch pilot at a Venlo, Netherlands warehouse into regular operations in 2026 after hitting unloading rates of 600 cartons an hour. Boston Dynamics positions Stretch as a force multiplier that cuts operational errors by 40 percent and lifts daily order-processing capacity 20 percent without adding headcount or shifts — metrics that set a benchmark any new entrant in warehouse and food-logistics automation must meet.

Tesla's Optimus program illustrates the difficulty of scaling a general-purpose humanoid from demo to production line. First unveiled in 2021, the program has seen repeated timeline revisions. The delay matters for Atoms because Optimus targets repetitive manufacturing and logistics tasks that Atoms' physical-AI platform also addresses in food and transport. Every quarter Optimus slips, the window widens for specialized, vertically integrated solutions to win contracts and build installed bases.

Foxconn, the world's largest contract electronics manufacturer, has moved to secure its position in the physical-AI supply chain. In 2025 the company announced a strategic partnership with SAP to co-innovate on Physical AI, SAP Supply Chain Management, and Foxconn Smart Manufacturing across the Asia-Pacific region. Chairman Young Liu called the partnership "an important step toward realizing the vision of intelligent and AI-powered manufacturing." For Atoms, which will need high-volume, reliable hardware production as it scales beyond pilots, Foxconn's deepening involvement in AI hardware assembly represents both a potential partner and a competitor that could integrate vertically into the same industrial-automation stack.

The competitive response is not limited to robotics pure-plays. Autonomous trucking startups continue to raise capital and lock in lane agreements. Atoms enters a market where incumbents have proven ROI data, installed bases, and multi-year service contracts. Its advantage, if it materializes, will come from a software-first architecture that treats food, mining, and transport as a single compute platform — rather than three separate automation projects. The next 12 months will test whether that architecture can convert pilot agreements into recurring revenue faster than the established players can extend their lead.

Sacramento and Washington Weigh In

California's manufacturing push runs on a simple premise: tax credits that pay out only after companies deliver jobs and capital investment. The California Competes Tax Credit program, administered by the Governor's Office of Business and Economic Development, has directed credits to more than 1,200 businesses since inception. Those recipients committed to nearly 160,000 full-time positions and over $50 billion in private investment across the state, per GO-Biz figures released in November 2025.

The latest round sharpened the focus. On November 19, Governor Gavin Newsom announced $99.9 million in CalCompetes awards to nine companies expanding headquarters and manufacturing operations. The package targets 2,752 new jobs with an average weighted annual salary of $139,000 and expects to unlock more than $370 million in fresh private capital. Manufacturing, the administration's Economic Blueprint labels the sector both a "strengthen" and "accelerate" priority, delivered $405.6 billion in output in 2024 and employed 1.24 million workers.

"These companies represent the next generation of California innovation," said Dee Dee Myers, Senior Advisor to Governor Newsom and Director of GO-Biz. "From advanced manufacturing to clean technology, these investments strengthen the industries that define our economy and accelerate inclusive, economic growth across the state."

The November cohort illustrates the program's reach.

Accountability runs both ways. The same November committee meeting that approved new awards also recaptured credits from 17 companies that missed milestones or exited agreements — including Better Holdco, Moxion Power, and Sorrento Therapeutics, each forfeiting $25 million. The recapture mechanism keeps the pool available for firms that execute.

Federal money layers on top. The CHIPS and Science Act of 2022 channels roughly $50 billion through the Department of Commerce to bolster domestic semiconductor capacity, advanced research, and supply-chain resilience. While Atoms builds physical AI systems rather than chips, its robotics stack depends on domestic compute and sensor supply chains that CHIPS funding aims to secure.

California's Jobs First Economic Blueprint, released in February 2025 and implemented by a nine-agency council, adds workforce training and regional development to the tax-credit framework. For a firm deploying autonomous robots across those verticals, the convergence of state credits, federal supply-chain investment, and a manufacturing workforce pipeline creates a tailwind that didn't exist five years ago. The next question is whether Atoms files for its own CalCompetes allocation — or lets rivals claim the credits first.

Kalanick's manifesto closed with a line about bleeding but not perishing, about fighting back into the arena. The arena is now public. Forty-two roles posted. Nine added in a week. Boston Dynamics building a $100 million factory in Waltham. Foxconn locking in AI hardware contracts. Caterpillar and Komatsu running hundreds of autonomous trucks. The physical-AI race has such a participant. The only question left is how fast the robots deploy.

Working in frontier tech? Zero G Talent tracks the openings: see every open Boston Dynamics role, browse frontier tech jobs, openings at Atoms, and the people building the field.